The six tests

INVEST reads as one word, but it is six independent yes/no questions. Asking them one at a time turns “not ready” into a precise verdict.

A story that answers Yes to all six is ready to hand to the Author; one No is a repair order, not a rejection. Switch the story on the right to score it against the six and read the pass count and the first failing letter.

The six questions One letter at a time On the Portal References

Score a Portal story against the six · switch the story

enrol in a course — passes 6 / 6. Ready to hand to the Author.

Six questions, one at a time

INVEST is one word, but reading it as one judgement loses its value. It is six separate questions, and a story is ready only when all six answer Yes:

Independent: Can this be built and shipped without first finishing another story?
Negotiable: Is the detail still open to a conversation, not frozen into a contract?
Valuable: Does it change what a user can do, demonstrably?
Estimable: Can the cost be guessed with any confidence?
Small: Does it fit one rung — one build-and-demo turn?
Testable: Is there a concrete way to confirm it is done?

The first No is the most useful output. A story that fails Small cannot be specified for one turn; one that fails Testable cannot be accepted. Reading the six in order finds the repair before the rung is built.

Each letter is a separate gate

The scorecard frames the whole verdict at once. This view proves the consequence underneath it: each letter is an independent gate, so a story can pass five and still be blocked by the sixth. Pick a letter to read its yes/no question, the Portal story that cleanly passes it, and the one that cleanly fails it.

Reading letter by letter is what keeps the repair precise. “Not ready” is a feeling; “fails Small” is an instruction. The inspector shows that the same Portal story can be a clean pass on one letter and the textbook failure on another — the letters do not move together.

Inspect one letter · its question, a pass, a fail

I — Independent: can it be built and shipped without finishing another story? PASS: open a lesson (waits on nothing). FAIL: manage the whole catalogue (its parts depend on one another).

Six gates, read one at a time: a story can pass five and still be stopped by the sixth, and the one it fails is the repair.

The six tests on the Portal

On the Portal — the learning platform this course builds — take the story “enrol in a course”. Read the six in order. Independent: it needs only a course id and a learner id, both already minted by Portal.ID.generate/1. Negotiable: the confirmation copy and the duplicate-enrol rule are still open. Valuable: the learner gains access they did not have. Estimable: one stored event plus a read check, a guessable cost. Small: one rung, one turn. Testable: after enrol, the learner's enrolment list contains the course. Six of six.

Now read “manage the whole catalogue”. Independent fails — its parts depend on one another. Estimable fails — no one can put a number on months of work. Small fails — it is not one rung. Three letters fail at once, because size drags the others down. The story is not wrong; it is not yet a story, and the first failing letter says to split it.

# Score "enrol in a course" — read the six in order.
#   I  independent  — needs a course id + a learner id, both already minted
#   N  negotiable   — confirmation copy and duplicate rule still open
#   V  valuable     — the learner gains access they did not have
#   E  estimable    — one stored event + a read check
#   S  small        — one rung, one turn
#   T  testable     — after enrol, the enrolment list contains the course
course_id = Portal.ID.generate(:course)   # => "CRS0Nb...", the ids the story rests on
# 6 / 6 — ready to hand to the Author.

# "manage the whole catalogue" — I, E, S all fail. Not yet a story; split it (A2.05).

The principle

Read the six tests one at a time; the first No is the most useful output, because it names the repair before the rung is built.

On the Portal

“Enrol in a course” scores 6/6 and is built; “manage the whole catalogue” fails I, E, and S, so the first No sends it back to be split.

A story ready to hand over passes all six; one that fails any letter is told, by that letter, what to repair.

Recap — six questions, one verdict

INVEST is six independent yes/no questions, read before a rung is built. A story that answers Yes to all six is ready to hand to the Author; one No is a repair order that names what to fix. Each letter is a separate gate, so a story can pass five and be stopped by the sixth. The next dive turns this scorecard into a diagnostic: when a story fails, it fails in one of a few recurring ways — a smell — and the smell names the letters and the rewrite.

Next in A2.03: A2.03.2 · Story smells — the four recurring ways a story fails, and how to diagnose and rewrite each.

References

The checklist's source and the story tradition behind it, and where it connects in the course.

Sources

Wake, B. — INVEST in Good Stories, and SMART Tasks — the six tests, in the words that coined them.
Cohn, M. — User Stories Applied — what a good story is, and how to keep it small.
Beck, K. — Extreme Programming Explained — stories and the customer's definition of done.

Related in this course

A2.03 · INVEST — the module hub.
A2.03.2 · Story smells — the next dive.
A2.03.3 · Small and independent — the Small/Independent tension.
A1.04 · Two layers: roadmap and specs — the spec a passing story feeds.
A2 · Decomposition — the chapter.
Companion · Functional Programming in Elixir — the Portal's internals.